Home Theater Forum and Systems banner

1 - 20 of 48 Posts

·
Registered
Joined
·
23 Posts
Discussion Starter #1 (Edited)
Hi,

This is my first post here and I'm hoping to get some advice regarding bass traps or acoustic treatments for my apartment. It's a rented space, and it's not very big, so I want them to be as unobtrusive as possible. My setup is 2-channel only and I've already purchased Jim Smith's book "Get Better Sound" and have achieved some benefits from his setup tips. I've also purchased a Behringer ECM8000 mic and have tested the room with REW. Based on the results I was able to further optimize my speaker and listening positions. I've attached a pic of the best graph I was able to achieve by moving the speakers and mic position, and I'm wondering if I can improve on this with some DIY bass traps.

I've been doing some thinking about it, and cutting out cardboard templates, and I think I could make some 24" tri-corner traps that would be fairly easy to install in the positions shown in the Sketchup drawing below. I was also thinking I might be able to put a soffit style panel across the upper rear wall over the desk, and perhaps install some downlights to dress it up a bit. I may also try building some bass-trap planter boxes similar to the ones offered by RealTraps.

Any thoughts on these ideas? Anything else I should consider since the room is L-shaped and my set up is asymmetrical? Right now I have blinds and thin curtains over the large windows, and I was also wondering if I would be better off starting with some heavy curtains or blinds instead of bass traps.

Thanks very much and any advice would be much appreciated.
 

Attachments

·
HTS Senior Moderator
Joined
·
5,288 Posts
While those locations will help address excess bass ringing in the room, in my opinion, the front right corner is a prime location for bass absorption. The right speaker is much more boundary and somewhat corner loaded than the left which is going to cause it to sound much richer and fuller than the other one. You'll also want to address the early reflections off that right wall which are going to be more intense and arrive earlier in time than the ones off the opposite wall.

Bryan
 

·
Registered
Joined
·
23 Posts
Discussion Starter #3
While those locations will help address excess bass ringing in the room, in my opinion, the front right corner is a prime location for bass absorption. The right speaker is much more boundary and somewhat corner loaded than the left which is going to cause it to sound much richer and fuller than the other one. You'll also want to address the early reflections off that right wall which are going to be more intense and arrive earlier in time than the ones off the opposite wall.

Bryan
Thanks for your response. Actually, my first thought was to fill that right corner with a SuperChunk or Studiotips Corner Absorber trap, but I thought it might be better to distribute the absorption around the room a bit. It sounds like it might be better to completely fill that right corner first.
 

·
HTS Senior Moderator
Joined
·
5,288 Posts
You will get some benefit in decay time by spreading things out. That said, that corner and the close right wall still need to be done. Then you can maybe do the upper rear corners.

Bryan
 

·
Registered
Joined
·
23 Posts
Discussion Starter #5
Thanks again for your advice. I had a mirror out last night and with my current setup the first reflection point on the right wall is in the seam between the door and the wall where the door is hinged. So treating that point is going be difficult. I may be able to move the speakers forward and my seat back to put the reflection point on the back of the door, but I don't think I have the room and I'm not sure I want to hang a panel on the door. Maybe I'll start with the right corner for now and see what happens.

Cheers,

Andrew
 

·
HTS Senior Moderator
Joined
·
5,288 Posts
Even without getting at the reflection area, having a panel directly beside the speaker will help minimize the uneven boundary loading.

Bryan
 

·
Registered
Joined
·
23 Posts
Discussion Starter #7
Even without getting at the reflection area, having a panel directly beside the speaker will help minimize the uneven boundary loading.

Bryan
That would be much easier, and not too obtrusive. I really don't want to fill the corner with a superchunk, but would be willing to put a tri-corner in the upper right, and a 2'x4' panel on the right wall beside the speaker. Would you recommend 4" thick with FRK facing out for both of them?
 

·
Banned
Joined
·
516 Posts
Brian is trying to help you, but I fear your concern for aesthetics has already significantly limited your options.

Let me summarize:

First, your choice of speaker placement lacks Left-Right symmetry.

The significance of this is that the imaging, localization is pretty much hosed unless you can effectively establish that.

Brian has politely suggested a partition and bass traps.

But realize that this reflective partition will necessarily need to be large. It must be large enough to be larger than the wavelengths down to about 400 Hz (at least 3 ft by 3 ft per each pathway) and large enough to cover the multiple reflective pathways affecting the seating area.

This is further complicated by the fact that you will have to make this boundary perform similarly to the other side boundary – while also surgically controlling the high gain, early arriving reflections. Thus you will be selectively applying absorption to the reflective boundary. A mirror is not sufficient to determine this complex amalgam of behaviors. You NEED to use the ETC response to identify specific reflections and to adjust gain to both effectively control the various reflections as well as to create acoustical symmetry. And don’t’ forget the ceiling!

As to a comment regarding your questions: A 2'x4' panel would not be large enough to control 'well controlled' distribution down to 400 Hz where the wavelength is ~2'8". And not having a reflective surface there is already equivalent to having an absorber there! You just have no idea regarding what the energy is doing and how it correlates to the other boundary surface! You need an ETC response for each speaker to discover what is actually happening! that will determine the exact nature of the boundary, be it reflective, absorptive or a combination of the two.

Much of this could be alleviated if the room were rearranged for optimal sound quality by relocating the speakers to the end of one of the alcoves, providing a degree of symmetry… And then using the ETC to surgically treat the side walls and ceiling as necessary.



The other major fly in to ointment is room modes. You have what is known as a coupled space. You will effectively have modes that are conditioned by 3 regions: Each of the side alcoves, and also by the entire space. Compounding this is the similarity in the dimensions of each space! They will sum. Prediction is beyond the already severally restricted generally available tools.

The only way to determine the distribution of these modes is to measure and map the space.
I appreciate your reluctance to employ obtrusive bass trapping. But as you are opting to go the porous route, you don’t have much choice. Tri- corner traps are very cool looking, but they will not be extremely effective, especially anticipating the degree of trapping you will require. Also, while a straddled panel may be more convenient to make, they are not as effective as SuperChunk style, where you benefit from more porous material. And you are going to need more coverage area to be effective.

One additional ‘trick’ you can use is to use adjoining spaces as bass ‘sinks’ and to, where possible, leave the doors open and allow the energy to be distributed into a larger volume. Just realize that in doing so this will modify the modal distribution as well, so you will need to explore these options and measure and map the space in order to determine the best options – hoping that such manipulation will not place your listening position in a null.

And again with regards to your question: Right now you have lots of uncontrolled reflections about which you have no idea what is happening. And considering the limited volume of bass traps, I would not worry about FRK facing at this point - especially as it is hard to predict just where and how such reflection would be directed and at what gain. Again, ETC measurements made with simple panel template mock-ups would provide much needed information that would quickly and accurately determine the effect of the use of a reflective FRK surface...

Again, I would suggest that you use the waterfall response and determine the precise behavior involved. And then use this as guide for determining the optimal treatment, rather than simply estimating you needs based upon aesthetic concerns without any idea of what are the real problems.

Dependent upon the results, you may want to seriously consider some of the tuned Modex modules (RPG), placed in high pressure regions determined by modal mapping of the space.

But, as always, the choices are yours. Simple measurements will reduce the range of reasonable options from 'unlimited', to a more manageable range conditioned by the actual behavior. If you need any assistance, please don’t hesitate to holler. (But also realize that positing ‘what ifs’ without actual measurements of actual behavior will necessarily receive an indeterminate answer!:ponder: Good luck!
 

·
Registered
Joined
·
23 Posts
Discussion Starter #9
Brian is trying to help you, but I fear your concern for aesthetics has already significantly limited your options.
I am very appreciative of Brian's assistance, and I certainly hope I didn't say anything that would suggest otherwise. Unfortunately, this is not a dedicated listening room or home studio, so I do have to take the aesthetics of any acoustic treatments into consideration. Due to the placement of doors and windows, and the lack of electrical outlets on some walls, I also have very few options for equipment placement. So it's certainly a less than ideal situation, but it's one I have to live with for now.

Brian has politely suggested a partition and bass traps

But realize that this reflective partition will necessarily need to be large. It must be large enough to be larger than the wavelengths down to about 400 Hz (at least 3 ft by 3 ft per each pathway) and large enough to cover the multiple reflective pathways affecting the seating area... A 2'x4' panel would not be large enough to control 'well controlled' distribution down to 400 Hz where the wavelength is ~2'8"
The wall beside the right speaker is 4' wide from the rear corner to the door. I understand it's almost impossible to give definitive answers, but should I consider covering most of that wall, if not the whole thing, with 4" thick panels?

I'm generally pretty happy with the sound I'm getting and the imaging, but sometimes I find the bass pretty boomy, which is why I started looking into bass traps. Adjusting my listening position and the speakers has made a difference, but I thought some room treatments would also be worth considering, as long as they didn't take over the apartment and make it look like a recording studio.

I definitely appreciate the advice both of you have provided and it's given me lots to think about. I was all set to rush out and buy some insulation a few days ago, but I think I'm going to slow down and spend some more time learning about the various options.

Thanks again,

Andrew
 

·
Banned
Joined
·
516 Posts
Please take the time to make a few measurements (waterfalls and ETCs).
This step can significantly bring what may seem to be a pretty overwhelming issue down to a manageable task.

These will tell you exactly what is going on and will provide an inventory of behaviors you would want to focus on treating.
Armed with such information, not only will the many possibilities be reduced to a few realities, the options for effectively treating them can be better assessed. Many of the most egregious issues can often be treated rather easily and surgically - without the need to rebuild the entire house!:laugh:

Then you can explore the best options for each, maximizing their effectiveness while complimenting your home to the greatest extent possible.

And the effects of incremental changes can be quickly and objectively assessed.

I know many are scared by the thought of measurements, but they are actually quite easily done and WILL make your options more defined.
 

·
Registered
Joined
·
23 Posts
Discussion Starter #11
Please take the time to make a few measurements (waterfalls and ETCs).
This step can significantly bring what may seem to be a pretty overwhelming issue down to a manageable task.
Thanks again for all the info. I had to do some searching to find out what an ETC was, but figured it out and understand now that it's part of the Impulse Response. So what you're saying is I should measure the Impulse Response for each speaker separately? Sorry if this sounds like a stupid question, but do I just do this with the balance control on my amp? Should I do the same for the waterfall measurement, or should that be done with both speakers on?
 

·
Banned
Joined
·
516 Posts
You do not adjust anything unusual with your amp.

You measure the test signal as generated by a program such as REW generated through each speaker separately.

Thus you have one sweep generated with the left speaker, and without moving the mic, another generated through the right speaker.

The goal is to obtain a response of the signal from one loudspeaker and to then 'observe' the energy propagation from that one loudspeaker within the room, and of its various pathways and incidence with object boundaries within the space, until it reaches the measurement mic.

I would suggest before doing either measurement that you establish and mark the precise position of the measurement mic capsule placed at the spot occupied by your head at the listening position, by establishing a secure plumb bob from the ceiling (you can pull it aside when not dong measurements). By having this reference point available, you will be able to make subsequent measurements at the exact same point allowing comparison of the various generations of measurements. (Also, you will not move the speakers)

(Don't worry about the exact position of your ears! This is an issue referred to as inter-aural cross correlation - IACC- and it is well understood and we need not worry about that here.)
 

·
Registered
Joined
·
23 Posts
Discussion Starter #13
Thanks, I'll give it a try tonight. I have done some measurements with REW and still have tape all over my floor marking the various measuring positions. Instead of hanging a plumb bob from the ceiling, I attached a string and small weight to the mic so I could position it over the tape on the floor and get repeatable results.
 

·
Registered
Joined
·
23 Posts
Discussion Starter #14
You do not adjust anything unusual with your amp.

You measure the test signal as generated by a program such as REW generated through each speaker separately.

Thus you have one sweep generated with the left speaker, and without moving the mic, another generated through the right speaker.
Okay, I've been trying to figure out how to send the test signal from REW to each speaker separately and I'm stumped. I've been reading the help files and searching the forums, and can't find any settings in REW that will do this. I did find a post on avsforum that said you should measure one speaker at a time by disconnecting all but one speaker. Is that the best way to do this, or am I missing something in REW?
 

·
HTS Senior Moderator
Joined
·
5,288 Posts
You're not missing anything. Just disconnect one speaker.

The results will give you insight into what's happening. That said, given the limitations, it's likely to be an exercise in frustration. Knowing what is going on and where it certainly a good thing but if you can't deal with it, then it's not really much use. It will simply prove to you what needs to be done.

The front corner and right wall WILL need to be addressed. The symmetry is what it is and this is the best way to address it. Given the limitations on available space, I would agree that covering more of that wall would be desirable.

These are common problems in real world, non-dedicated spaces and we just try to address them the best we can given the limitations.

Bryan
 

·
Registered
Joined
·
23 Posts
Discussion Starter #16
You're not missing anything. Just disconnect one speaker.
Thanks for the clarification. You're probably right about it being an exercise in frustration, but I have the gear to do the testing and I am keen to understand what the sound "looks" like in my room.
 

·
HTS Senior Moderator
Joined
·
5,288 Posts
Don't get me wrong. Measuring the room is certainly worth doing. Even if you can't treat everything, it can show you how potentially small changes in speaker and seating position can improve frequency response.

Bryan
 

·
Registered
Joined
·
23 Posts
Discussion Starter #18 (Edited)
I ran some more tests tonight and here are some waterfalls and ETCs for each speaker. I'm trying to figure out how to interpret the results, and I'm not sure if I did them correctly, so right now they don't mean much to me. Any feedback would be much appreciated and I will keep reading and trying to learn more.

Thanks.

SPL Right Channel


SPL Left Channel


Waterfall Right Channel


Waterfall Left Channel


Impulse Right Channel



Impulse Left Channel

 

·
Banned
Joined
·
516 Posts
Can you post the raw response (wav or the native REW) files so that we can open it and adjust the windowing?

Just a bit of insight, Sound travels approx. 1.13ft/ms, thus displaying an ETC window to 1sec captures the direct and indirect signal travel up to 113 feet. My guess is that there is no appreciable specular reflections after about maybe ~50ms which would encompass energy traveling ~56.5 feet (more than twice the length of the room...). Thus beginning with a horizontal/X-axis time window of ~50 ms should be adequate to get a good overview - and one can zoom in from there as necessary for more detail.

Also, I would estimate that one would not need more than a window of about 40 dBfs on the vertical gain scale will be needed displaying from a little above 0dbfs to ~-40 dBfs.

Are you using loopback device correction to address hardware propagation time delay? (While necessary, it may be a moot issue as configured as I am not sure if we are able to see the total time of flight from speaker source to measuring mic.)*

Also, I fear a critical bit of data is not displayed.

While it may seem advantageous to translate the time axis and to assign t=0 to (generally) correlate to the arrival of the direct signal, we do not have the total time of flight from the source speaker to the measurement mic in order to determine the base time, and hence distance the sound traveled before being received at the mic. Thus, all we can do with what is displayed is to determine the difference in time and distance of each indirect signal relative to the direct signal.

While translation of the origin on the time axis is very useful for adjusting delay lines in multiple unit arrays and distributed systems, it is not useful for determining the distance and direction of the specular energy.

Thus, it would help if you would carefully and as precisely as you can, measure the distance from the center of the speaker baffle (assuming that is somewhat close to the acoustic center of the unit) to the mic capsule. We will need to adjust the distance calculations obtained from the ETC plot by that distance. The reason for this is that the signal does not travel from the point starting at the arrival of the direct signal to its arrival time (yeah, I know that sounds a bit absurd!), but instead the signal for all direct and indirect signals travels from the source speaker to the measurement mic, and thus in order to demonstrate how one can determine the precise actual distance of travel and the location of incident boundaries, we will need to know the complete information describing the total time/distance of flight.

After one understands this concept, it is not too difficult to explain a few other techniques to determine the precise path of travel.

(And if there is a way to overcome this in REW, I would greatly appreciate knowing it!)

With the raw data that we can massage, we can illustrate, explain, and suggest a few ways to make very practical use of the ETC response.

------------------------------

An aside:

*My request for a new feature would be to make the total time of flight the default time view (assuming the required loopback compensation is done, unlike gain and freq calibration which are optional, time calibration IS necessary!!!) Then one can then translate the T=0 time value origin for the limited cases where such translation is of practical use. If this issue were addressed (or at least made the default view if it is possible), then just about all of my other concerns would simply be issues regarding bells and whistles.

Also, a selectable dB SPL scale would be of MUCH more use than a dBfs scale which, while very useful for setting digital signal line levels, is not of use for sound pressure level measurements. And, again, as we are concerned with relative differences in gain and not absolute gain, gain calibration is not necessary.

For this curious about gain calibration (and the notion that many seem to feel it necessary), gain calibration is only necessary if one is conducting environment certified noise level analysis (NLA) measurements. And if that is the case, the entire rig will need to be certified and calibrated -something that is far outside the scope of what is possible here. The other use would be if more than one response made in different locations/environments were being compared there was required an objective reference baseline for comparison, but one would be hard pressed to imagine such a practical scenario where such would be the case.

--------------------------------------------

As far as the modes are concerned, I have quite a few questions and there is likely more work to be done.
I assume the waterfalls were taken with the mic in the listening position.

The problem with this is that you are most likely in the nearfield of he speakers and you will be dealing not only with a measurement of the modal support of the room, but of a high level direct signal of the speakers.

In order to determine the lowest supported mode of the room, the easiest way to do this is to place a speaker in the front floor corner of the room, and then to place the mic on a tall stand facing the rear diagonally opposite ceiling corner

And you may also be interested in the cancellation of signal due to superposition (folks need to learn this term!) that refers to how waves sum when multiple sources are used, and also when virtual sources in the form of reflections are manifest, creating spatial polar lobing. This polar lobing consisting or regions of reinforced response alternating with regions of phase cancellation, will appear to vary with frequency depending upon where you are located in the field. It is this polar lobing that is responsible for the pattern seen in a frequency response called comb filtering. It is important to understand that there is no such 'thing' as comb filtering (it is just a pattern in a display!). The real 'thing' that actually exists is the behavior known as spatially distributed polar lobing caused by the combination of two spaced real or virtual sources. And this will contribute to the modal measurements made at the listening position when more than one LF source is driven, as well as by possible contributions by virtual sources (boundary reflections) called SBIR.

Thus a few related variables will need to be systematically researched and identified and/or eliminated as contributory variables.

Its not hard, but a few well thought out steps may be necessary to isolate the specific causes contributing to the total bass response. Thus, measurements will tell you what is happening, but you will need to think a bit in order to apply an understanding of the various acoustical physics concepts that can be at play depending upon the setup. The reason is that if the behavior that you are viewing is a result of multiple causes, you need to identify the particular cause that effects each 'part', and correct that. One solution will nor magically correct issues caused by multiple causes. Thus, this is not a limitation of the measurements! It simply requires that you be at least as smart as your tools that can expose such behavior. And depending on the circumstances, you may need to conduct a few small experiments to isolate each contributory cause so that they can be corrected.
 
1 - 20 of 48 Posts
Top